Potential mediators of the modest association between retrospectively rated parental behaviour and adult affective symptoms in offspring were investigated in a national longitudinal study of a cohort followed to the age of 43. Personality measures from adolescence could account for a small part of this association. Personal relationships in adulthood were more strongly associated with both parental behaviour and symptoms: marital history, emotional support, social network and availability of help in a crisis. Poor parenting did not lead to a general vulnerability to later life events, and socio-economic status and financial hardship were not implicated in the link between parental behaviour and adult symptoms. However, parental affectionless control was associated with certain types of life stressors in adulthood, i.e. interpersonal as opposed to non-interpersonal life events. Collectively, aspects of personal relationships accounted for much of the elevated symptom levels in those rating parents as low on care or high on control. Findings were consistent with the notion that interpersonal competence is important in the continuity between childhood experience and adult mental health, but other possible interpretations are discussed.

This paper reviews the evidence for changing patterns of mental health over time in childhood and youth in Western societies. The evidence suggests that the prevalence of major depression, substance abuse and offending behaviour, as well as the incidence of suicide, is increasing in adolescent populations, particularly among males. There are also indications that problem behaviours among younger children are becoming more common. There is no evidence of a deterioration in the adjustment of the pre-school population.

Long-stay patients who were discharged from Friern and Claybury hospitals during the 6-year period between September 1985 and August 1991 were followed up for at least 1 year and the details of all readmissions were recorded. Of the 615 hospital leavers, some 178 (29%) were readmitted at some time during the follow-up period. A proportional hazards regression analysis revealed a number of independent baseline factors which affect the risk of readmission at any time during the follow-up period. Younger patients were more at risk of readmission than were older patients and ‘new’ long-stay patients had a greater risk than baseline patients. Risk was also greater in patients who had a larger number of pre-discharge admissions,a higher abnormal behaviour score (PSE BSO) or a diagnosis of manic psychosis. An ordinal logistic regression analysis revealed that these same five factors tended to increase the actual number of readmission episodes. The proportion of post-discharge time spent in hospital among the readmitted patients was found to be greater for males, new long-stay patients and for patients with a higher number of positive social behaviour symptoms (SBSPOS score).

We evaluate the long-term test–retest reliability and procedural validity of phobia diagnoses in the UM-CIDI, the version of the Composite International Diagnostic Interview, used in the US National Co-morbidity Survey (NCS) and a number of other ongoing large-scale epidemiological surveys. Test–retest reliabilities of lifetime diagnoses of simple phobia, social phobia, and agoraphobia over a period between 16 and 34 months were K = 0·46, 0·47, and 0·63, respectively. Concordances with the Structured Clinical Interview for DSM-III-R (SCID) were K = 0·45, 0·62, and 0·63, respectively. Diagnostic discrepancies with the SCID were due to the UM-CIDI under-diagnosing. Post hoc analysis demonstrated that modification of UM-CIDI coding rules could dramatically improve cross-sectional procedural validity for both simple phobia (K = 0·57) and social phobia (K = 0·95). Based on these results, it seems likely that future modification of CIDI questions and coding rules could lead to substantial improvements in diagnostic validity.

Childhood parental loss may be an important risk factor for psychiatric illness in adulthood. While this association has been carefully examined for depression, little is known about the role of parental loss in predisposing to alcoholism. We examined an epidemiological sample of female twin pairs with the same history of continuity or disruption in parent–child relationships (N = 1018 pairs; mean age 30 years), using a range of definitions of alcoholism. Childhood parental loss through separation, but not death, substantially increased the risk in adulthood for all definitions of alcoholism. Furthermore, both paternal and maternal alcoholism substantially increased the probability of parental separation from their children. Proposing a structural equation twin-family model that incorporates childhood parental loss as a specified environmental risk factor, we examined how much of the association between childhood parental loss and alcoholism was causal (i.e. mediated by environmental factors) v. non-causal (mediated by genetic factors, with parental loss serving as an index of parental genetic susceptibility to alcoholism). Both the causal and non-causal paths were significant for all definitions of alcoholism. However, the causal–environmental pathway consistently accounted for most of the association. While a significant proportion of the association is due to non-causal genetic mechanisms, childhood parental loss (or the familial discord that precedes or follows it) is probably a direct and significant environmental risk factor for the development of alcoholism in women.

This study aimed to compare the costs of treatment by community-based and hospitalbased psychiatric services. The design entailed random allocation of patients presenting with psychiatric emergencies over a subsequent 3-month period to one of two services, followed by retrospective quantification of service use and its cost for each group. One hundred patients with emergency presentations to the psychiatric service via the Accident and Emergency Department, liaison psychiatrist and approved social worker were included in the study. Their use of a range of items of service was recorded and disaggregated costings of these items of service was calculated. The use of non-psychiatric services was similar for both groups, but the use of psychiatric services differed, with the hospital group making greater use of in-patient beds and the community group employing more frequent home-based interventions. The total cost of treatment for the community group (£56000) was much lower than for the hospital group (£130000), although the median patient cost was 50% higher in the community group (£938 v. £610), and a greater proportion of the community service expenditure (10% v. 2%) was due to failed contacts. Taken together with clinical outcome, which showed no advantages for the hospital-based service over the community-based service, our findings suggest that this form of community psychiatric service is a cost-efficient alternative to hospital-based care for this group of patients.

Studies examining a possible decline in the incidence of schizophrenia over the last two to three decades have paid little attention to the possible role of birth cohort effects. We collected data on a Scottish national sample of all schizophrenic patients, admitted for the first time between 1966 and 1990 (N = 11348; male = 6301). In an Age–Period–Cohort analysis, a full model, incorporating three factors, had a substantially better fit to the data than other models (especially, an Age–Period model), providing clear evidence of the presence of a cohort effect. After adjustment for the effects of age and period, there was a 55% reduction in the rate of schizophrenia in men and a 39% fall in the number of women over the 50-year birth period from 1923 to 1973. The marked decline in the first admission rates observed in Scotland cannot, however, be attributed entirely to this cohort effect. Rather, a greater proportion of the declining first admission rates (88%) is ascribed to the period effect (i.e. artefactual or causally related cross-sectional effects). Nevertheless, the fact that a birth-cohort effect accounts for part of the declining incidence, suggests that causal environmental factors operating early in life have been diminishing in intensity.

African-Caribbean (N = 136) and White British (N = 192) female family planning clinic attenders were administered the Bulimic Investigatory Test, Edinburgh (BITE) and the General Health Questionnaire (GHQ-28). A proportion of the participants were subsequently interviewed. The African-Caribbeans were found to have both significantly more disordered eating attitudes and a significantly higher level of abnormal eating behaviour than the White British. Although the African-Caribbean group had a significantly higher mean Body Mass Index this did not mediate the difference in levels of eating attitudes. When compared with the White British group more African-Caribbean women reported feelings of failure, guilt, abnormality and self-consciousness concerning their eating habits. The results indicate that eating problems may be highly prevalent in this ethnic minority population and suggest that there may be differences in the nature of eating disorder psychopathology between ethnic groups.

Coping has attracted much attention in research as a possible mediator of the psychological impact of cancer. Yet, conceptual ambiguity and methodological limitations have resulted in weak and contradictory findings. A major shortcoming has been the use of designs which do not represent adequately the complexity or the diversity of demands which arise from the diagnosis and treatment of cancer. The neglect of appraisal in the assessment of the relationship between coping and mental health is of particular concern, given the role it has been found to play in the onset and maintenance of affective disorders.

In a prospective study 673 newly diagnosed cancer patients were interviewed 4 to 8 weeks and 1 year later to assess the effects of their appraisals, coping responses and resolution of any concerns on subsequent mental health. Logistic regression analyses, adjusted for possible confounding variables, were used to investigate the relationships between coping variables and affective disorders. When examined separately the degree of threat appraised, the reporting of a helpless response and perceived success of primary responses in resolving concerns all predicted subsequent affective disorder. However, in a multivariate model only appraisal and success of the response in resolving the concern were significant. No response was found to prevent affective disorders, however certain types of responses were associated with the resolution of specific concerns. The results highlight the importance of assessing the whole coping process and the need to address the complex and multifaceted characteristics of cancer demands. On the basis of our findings we describe a maladaptive cycle of coping, which we believe contributes to the later onset of affective disorders in cancer patients.

This analysis examines some of the psychological, philosophical and sociological motives behind the development of pauper lunatic asylum architecture in Ireland during the time of the Anglo–Irish union (1801–1922). Ground plans and structural features are used to define five psycho-architectonic generations. While isolation and classification were the prime objectives in the first public asylum in Ireland (1810–1814), a combination of the ideas of a psychological, ‘moral’, management and ‘panoptic’ architecture led to a radial institutional design during the next phase of construction (1817–1835). The asylums of the third generation (1845–1855) lacked ‘panoptic’ features but they were still intended to allow a proper ‘moral’ management of the inmates, and to create a therapeutic family environment. By the time the institutions of the fourth epoch were erected (1862–1869) the ‘moral’ treatment approach had been given up, and asylums were built to allow a psychological management by ‘association’. The last institutions (1894–1922) built before Ireland's acquisition of Dominion status (1922) were intended to foster the development of a curative society.

Growth hormone (GH) responses to growth hormone releasing hormone (GHRH) of 53 in-patients meeting DSM-III-R criteria for major depressive episode with melancholia (24 non-delusional and 23 delusional depression) were compared with those of 19 healthy controls. No significant differences in basal GH were found between the control and either the non-delusional or the delusional groups. The whole group of depressed patients showed a significantly lower response than the control patients at all points of the GH response to GHRH curve as well as a lower area under curve. When the three groups (control, delusional and non-delusional depressed) were compared, it was found that only the non-delusional depressed patients had a significantly lower area under curve and lower values at +60, +90 and +120 min than the controls. The only significant difference between the two groups of depressed patients was that the delusional group showed a delayed appearance of the maximum response peak and a more prolonged response.

Mentalizing ability was studied in 46 symptomatic schizophrenic patients and 44 non-symptomatic controls. Subjects heard six stories and simultaneously were shown simple cartoon pictures depicting the action sequencing occurring in the stories. All the stories involved false belief or deception, so that it was necessary to infer the mental states of the characters in order to understand their behaviour. After each story, subjects were asked one memory/reality question (concerning an event in the story) and one question that depended on the ability to infer the mental state of one of the characters. Patients with paranoid delusions were impaired on the questions concerning mental states. Patients with behavioural signs (i.e. negative features or incoherence) were also impaired on the mental state questions, but this difficulty was associated with memory impairments. Patients with symptoms of passivity (e.g. delusions of control) and patients in remission did not differ from normal controls. These results are consistent with the hypothesis that certain of the positive symptoms of schizophrenia reflect an impairment in the ability to infer the mental states of others.

The CANTAB battery of neuropsychological tests was used to compare the performance of 28 patients with unipolar depression with that of 22 age and IQ matched controls. The patients were impaired on almost all tests studied with deficits in pattern and spatial recognition memory, matching to sample, spatial span, spatial working memory and planning. Most of the patients showed at least some impairment and deficits were seen across cognitive domains. An important finding was the detrimental effect of failure on subsequent performance; having solved one problem incorrectly, patients were far more likely than controls to fail the subsequent problem. Superimposed on the general deficits, there were also specific deficits in executive tasks characteristic of frontostriatal dysfunction and deficits in mnemonic tasks characteristic of temporal lobe dysfunction. This combination of a specific form of motivational deficit, resulting in oversensitivity to negative feedback, and superimposed specific neuropsychological deficits were correlated with severity of depression. The most significant correlations were seen between mnemonic deficits and clinical rating scores. Comparisons of the deficits seen in the depressed patients in this study with other patient groups assessed with the CANTAB neuropsychological battery, showed that one of the hypotheses of the neuropsychological deficits in depression, that of ‘frontosubcortical’ or ‘frontostriatal’ dysfunction, was not supported. These findings are discussed in relation to the likely neural substrates of depression.

From 1983–1994 a community mental health programme was set up in Guinea-Bissau. The first part of the programme concentrated on epidemiological aspects: rural and urban study areas were selected on socio-economic level and participation in the liberation war. A two-stage design was used to screen 351 adult consecutive general health care attenders and 100 children in a rural and an urban area for mental disorder. Psychiatric disorders have a morbidity of 12% among adults seen in Primary Health Care. Disorders were mainly neuroses (74%), but more psychoses were found than in other countries. No statistically significant difference in morbidity was found between rural-urban areas or between previous war and non-war zones. The diagnostic sensitivity of the Primary Health Care workers was 31%, their diagnostic specificity 88%. Thirteen per cent of the children showed neuropsychiatric disturbances. There were no sociocultural impediments to this public mental health approach. During the following intervention programme 850 Primary Health Care workers were trained and supervised nationwide. The diagnostic sensitivity of major mental disorders and epilepsy increased from 31% to an average of 85%. Before the training, their knowledge of the treatment of these disorders was nil whereas after training 82% of the patients received appropriate treatment. Moreover, this model programme shows a profitable cost/benefit ratio and a high sustainability over the last 10 years.

We used survival analysis to test for a suspected effect of co-morbidity of DIS/DSM-III major depression and panic attack on the age of onset of each of these disorders, in a population-based sample from four sites of the National Institute of Mental Health (NIMH) Epidemiologic Catchment Area Program (N = 12668). The possible effects of gender and presence of a co-morbid disorder on age of onset of the single disorders were analysed using the Cox proportion-hazard model, with gender as an explanatory variable and the co-morbid disorders as time-dependent covariates. The disorders begin earlier in life when they are co-morbid than when they occur singly. The occurrence of one disorder increases the chance of occurrence of the other. Females have from 1·7 to almost twice the risk for developing panic and/or depression compared with men, even if the co-morbid disorder precedes the index disorder or is lifetime co-morbid. These findings are compatible with a hypothesis of a higher severity in co-morbid cases when compared with a single disorder.

The question addressed to in this paper is whether severely mentally ill patients are treated differently in a community mental health service without the back-up of a mental hospital (South-Verona, Italy) compared with an institution-based system in which mental hospitals, although highly modernized, are still predominant (Groningen, The Netherlands). Using the psychiatric case-registers in both areas, the patterns of care in 2 years of follow-up of schizophrenic patients were constructed. Survival analysis was used to analyse in-, day- and out-patient episodes of care. Three-quarters of the Groningen and half of the South-Verona patients experienced at least one episode of hospitalization; 20% of the Groningen and 5% of the South-Verona patients were long-stay patients at the end of the observation period. The South-Verona patients had more episodes of in-patient and especially of day-patient and out-patient care. Cox's regression showed that the duration of episodes controlled for the history of events and sociodemographic characteristics, was significantly shorter in South-Verona. One of the main conclusion was that hospitalizations for the severely mental ill are also needed in a community-based system of care, supporting the assumption of a ‘bed-rock’ of mental illness. However, the South-Verona community mental health service seems to be able to reduce the duration of hospitalizations considerably.

The aim of the study was to investigate the recovery and frequency of physician contact in patients with unexplained physical symptoms and to identify factors associated with persistent disorder. Of 100 consecutive patients who presented with medically unexplained symptoms to a general medical out-patient clinic, 81 participated in a follow-up study. The mean follow-up time was 15·2 months (S.D. 4·0). At follow-up, many of the patients with unexplained physical symptoms reported that they had recovered (30%) or improved (46%) with regard to their presenting symptoms. Female sex and a high number of symptoms predicted a bad outcome in terms of recovery. Persistence of symptoms was not related to the duration of the symptoms, type of presenting complaint or the presence of psychiatric disorder. Forty per cent of patients with unexplained symptoms did not visit their general practitioner on their own initiative in the year following the initial visit to the clinic. Medical care utilization appeared to be associated with female sex, age, number and duration of symptoms, fatigue and psychiatric disorder, especially somatoform disorders. However, the association of a high frequency of physican contact with female sex and psychiatric disorder was not sustained after controlling for possible confounding factors.

This paper describes the construction, refinement and implementation of a self-administered measure of personal hopefulness, the Hunter Opinions and Personal Expectations Scale (HOPES). Initial state and trait versions of the HOPES instrument were utilized in three separate studies, comprising a medical student sample (N = 211), an adolescent male sample (N = 280) and a psychiatric hospital staff sample (N = 318). A revised 20-item, two factor, trait version of the scale was then utilized in a prospective, longitudinal investigation (N = 753) of the psychosocial sequelae of the earthquake which struck Newcastle (Australia) in December, 1989. Data from all four studies provide strong support for the HOPES instrument's construct, concurrent and predictive validity. Global personal hopefulness (GPH) was shown to be an enduring characteristic of individuals, with a test–retest correlation of r = +0·71 (over 64 weeks). The association between GPH and trait anxiety (r = −0·64) raised the possibility of redefining anxiety as hope under threat. The hope subscale (HS) and the despair subscale (DS) were moderately negatively correlated (r = −0·32), suggesting that hope and despair are not simply polar opposites. There were no gender differences in GPH scores, however, there were relatively clear age effects, with those aged 70 years and over reporting the lowest levels of personal hopefulness. GPH was negatively correlated with post-earthquake scores on the General Health Questionnaire (r = −0·33), the Impact of Event Scale (r = −0·33), the Beck Depression Inventory (r = −0·54) and the global symptom index from the SCL-90-R (r = −0·43). Overall, the contribution made by personal hopefulness to post-earthquake morbidity was equal to the contributions made by initial exposure to disruption and threat experiences.

Groups of home and school situational hyperactive primary schoolboys identified from the community were compared with pervasive hyperactive and non-hyperactive controls on a wide range of measures. The hyperactive groups tended to persist in the same category over a half-year period. Both situational hyperactive groups had lower measured activity levels than the pervasive hyperactive group and only the latter differed from non-hyperactive controls. Home hyperactivity was characterized by poor family relationships and was not distinguishable from non-hyperactive home-antisocial controls. School hyperactive boys had specific correlates of low intelligence, motor clumsiness, poor reading and academic abilities. Pervasive hyperactive subjects differed from both situational groups in showing a higher percentage of delayed language development. While home hyperactivity has dubious identity, the distinct pattern of external correlates in school and pervasive hyperactivity speak for the need to regard these as separate entities.

We studied the prolactin and cortisol responses to the 5-HT releasing agent, D-fenfluramine in 31 drug-free depressed patients and 29 healthy controls, using a double-blind, placebo-controlled design. There was no difference in either endocrine response between depressives and controls. Examining the sexes separately, or restricting comparisons to patients with melancholic depression, did not lead to any differences between the groups. Our findings suggest that in contrast to other 5-HT neuroendocrine probes such as L-tryptophan and clomipramine, the prolactin response to fenfluramine are not consistently blunted in depressed patients.